I found this very nice JMLR paper that seems to address exactly the problem I was raising: "How to Explain Individual Classification Decisions" [0]. They provide a general framework allowing to analyse the individual decisions of any classifier in terms of an "explanation vector", which corresponds to the local gradient of the class probability function: the value of its component d corresponds to the influence of the feature d with respect to the chosen class. Given its generality, I think it would make a very useful addition to the sklearn framework (although for my taste, it is a bit heavy on the math side). What do you think?
[0] http://jmlr.csail.mit.edu/papers/volume11/baehrens10a/baehrens10a.pdf On 2 October 2012 14:34, Christian Jauvin <[email protected]> wrote: >> * "Advice for applying Machine Learning" [1] gives general recommendations >> on how >> to diagnose trained models > > Thanks Immanuel, I find this document in particular to be a great > source of very practical advices and ideas. > > Christian ------------------------------------------------------------------------------ Don't let slow site performance ruin your business. Deploy New Relic APM Deploy New Relic app performance management and know exactly what is happening inside your Ruby, Python, PHP, Java, and .NET app Try New Relic at no cost today and get our sweet Data Nerd shirt too! http://p.sf.net/sfu/newrelic-dev2dev _______________________________________________ Scikit-learn-general mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
